Backfill Fetch

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

With no annotations, the description carries the full burden and does so unusually well: it discloses that every extraction is a paid call, the exact pricing ($0.15 fresh / $0.05 cached), that MCP calls return an error directing to the HTTP API, and that payment is never attempted over MCP. It also explains the fetch escalation path for bot walls and JS shells.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Is the description appropriately sized, front-loaded, and free of redundancy?

Three dense sentences with the purpose front-loaded and no filler. The payment and error-redirect details are essential, though the pricing parenthetical is slightly heavy for a tool that will always error over MCP.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a tool whose actual MCP behavior is 'always error and redirect to the paid HTTP endpoint,' the description covers what an agent needs to know: cost, endpoint, and that payment won't happen over MCP. The only gap is the unexplained 'mode' enum, which matters since no output schema exists.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is only 33%. The description hints at the 'auto' mode's escalation behavior but never explains the 'plain' vs 'render' enum values, and adds nothing about the 'url' parameter. The 'schema' parameter is documented in the schema itself, but the mode parameter remains underspecified across both fields.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Does the description clearly state what the tool does and how it differs from similar tools?

States a specific verb and resource ('Extract structured data from a web page') plus the output contract ('JSON matching a JSON Schema you supply') and names the backing service. An agent immediately knows what this does and that it differs from a raw fetch.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description explains the internal fetch strategy and the payment requirement, which implies the use case, but it never names or contrasts the 'fetch_page' sibling or states when to prefer it. Usage is inferable rather than explicit.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

This score is high because the description handles the entire behavioral profile in detail, covering the fetch escalation process (plain fetch first, then bot wall/JS shell), the pricing model, and the mandated failure path (external paid API link). Since no annotations exist, this detail is critical context that would otherwise be missing.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is highly dense and verbose, but its information is logically grouped: Definition -> Mechanics -> Cost/Error Handling. Every long sentence serves a necessary purpose, making its length justifiable.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

Given the complexity of the service (dynamic fetching, escalation, external payment processing), the description is extremely complete. It successfully explains the service's operational lifecycle and its failure modes, which is essential for this type of external tool.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 0%, meaning the description must define the parameters. The description only refers to the general action of fetching; it fails to explain the semantics or specific impact of the `mode` enum (auto, plain, render) or the required `url` format beyond generic fetching context.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Does the description clearly state what the tool does and how it differs from similar tools?

The description clearly defines the function as fetching a web page and returning readable markdown. While it doesn't explicitly detail how this differs from `extract_data`, the functional distinction (fetching vs. content extraction) is implied and well-established.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Does the description explain when to use this tool, when not to, or what alternatives exist?

It provides excellent operational guidelines regarding when the call is free (cache hits/daily allowance) versus when it requires payment and fails (402 error). However, it gives minimal guidance on when to use this tool relative to other fetching tools, offering no specific criteria for selecting it over alternatives like `extract_data`.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.